- From: Martin J. Duerst <mduerst@ifi.unizh.ch>
- Date: Wed, 11 Dec 1996 17:00:14 +0100 (MET)
- To: Erik van der Poel <erik@netscape.com>
- cc: Larry Masinter <masinter@parc.xerox.com>, Drazen.Kacar@public.srce.hr, J.Larmouth@iti.salford.ac.uk, www-international@w3.org
On Fri, 6 Dec 1996, Erik van der Poel wrote: > Larry Masinter wrote: > > > > If you use ENCTYPE=application/form-data, the form data from a POST > > can be labelled correctly. > > > > For POST with ENCTYPE=application/x-www-form-urlencoded, there is no > > simple way to label the charset. I propose that HTML be modified so > > that the very special construct: > > > > <INPUT TYPE=hidden NAME=charset> > > > > i.e., a form with a hidden field with the field name "charset" be > > interpreted as "supply the charset used for encoding the values". > > How about the ACCEPT-CHARSET parameter mentioned in the HTML i18n draft? > Can this be used by the server to indicate that the client should label > the charset in the urlencoded response? A discussion with Francois some evening in Sevilla revealed that probably the main function of the ACCEPT-CHARSET attribute on fields will be to smoth the transition to UTF-8 in URLs (query part in a GET request). Regards, Martin.
Received on Wednesday, 11 December 1996 11:02:33 UTC